The long-term objective of the proposed research is to identify the segments of the frequency spectrum which provide information on the location of a sound source. The specific aim is to test the hypothesis that monaural localization of sound is governed primarily by the "covert" peaks in the sound's spectrum. A covert peak is defined as that frequency segment within a given sound that is amplified more when the sound originates from one specific position in the horizontal plane - sagittal plane matrix than when it originates from any other position. Head related transfer functions, measured by a miniature microphone placed in the external ear canal, will provide the spectral data for the determination of covert peaks. The peak is termed "covert" because it may not appear as the frequency region of maximum amplification when dB gain is plotted against location of the sound source. It is hypothesized that no more than two covert peaks are required for monaurally localizing sound; that covert peaks also play a role in binaural localization; and, that all other characteristics of the sound's spectrum contribute minimally to the localization process. Normal listeners, rendered monaural by blocking one ear, will be required to locate broadband noise originating from among 91 loudspeakers positioned in the right auditory hemifield. Localization proficiency will be measured. Next, head related transfer functions for 1.0 kHz wide nose bands centered from 4.0 through 14.0 kHz in steps of .5 kHz will be obtained as these differently-centered noise bands emanate from each of the loudspeakers. From these data, covert peaks associated with the sound's spectrum for each loudspeaker position will be extracted. Those peaks peculiar to each location will be presented as stimuli to be localized. Localization proficiency will be compared to that recorded from a broadband noise. The adequacy of the covert peaks in guiding monaural location judgments will be evaluated. Binaural localization of sagittally positioned sounds will be also investigated with the aim of determining the relative contribution of covert peaks to the binaural process. Overall, these data will have relevance to (1) the assignment of hard-of-hearing people in the work place; (2) auditory spatial cues for the visually impaired; and, (3) the basic spectral requirements demanded of cochlear implants in order to reinstate spatial hearing.